MyCaffe
1.12.2.41
Deep learning software for Windows C# programmers.
|
The SolverInfo defines the user supplied arguments passed to each Worker. More...
Public Member Functions | |
SolverInfo (Solver< T > rank0, long hSrcKernel, long hSrcNccl, int nSolverRank, int nIterationOverride, string strCudaPath, List< ManualResetEvent > rgGradientReadyEvents, ManualResetEvent evtAllCreated) | |
The SolverInfo constructor. More... | |
Properties | |
Solver< T > | Rank0 [get] |
Returns rank Solver that will run in the Worker. More... | |
string | CudaPath [get] |
Returns the file path to the low-level CudaDnnDll.DLL file to use. Note, when null or emtpy, the path of the executing Assembly is used. More... | |
int | IterationOverride [get] |
Returns the training iteration override to use. More... | |
long | KernelHandle [get] |
Returns a handle to the kernel where the NCCL for this Solver was created (typically this is the kernel that also created the root Solver) More... | |
long | NcclHandle [get] |
Returns the handle to the NCCL instance for this Solver (typically this is created on the kernel that also created the root Solver, and must be transferred to the kernel of the CudaDnn instance running in the Worker). More... | |
int | SolverRank [get] |
Returns the rank of this Solver. More... | |
ManualResetEvent | InitializedEvent [get] |
Returns the event that is set after the Worker has completed initializing. More... | |
ManualResetEvent | StartedEvent [get] |
Returns the event that is set after the Worker has started running. More... | |
ManualResetEvent | AllCreatedEvent [get] |
Returns the event that is set after all Workers have been created. More... | |
List< ManualResetEvent > | GradientReadyEvents [get] |
Returns the event that is set after the gradients of the Solver in this Worker are ready. More... | |
Exception | Error [getset] |
Returns the error (if any) that occured when running the solver thread. More... | |
AutoResetEvent | ErrorEvent [get] |
Returns the event that is set when an error occurs. More... | |
The SolverInfo defines the user supplied arguments passed to each Worker.
T | Specifies the base type float or double. Using float is recommended to conserve GPU memory. |
Definition at line 519 of file Parallel.cs.
MyCaffe.common.SolverInfo< T >.SolverInfo | ( | Solver< T > | rank0, |
long | hSrcKernel, | ||
long | hSrcNccl, | ||
int | nSolverRank, | ||
int | nIterationOverride, | ||
string | strCudaPath, | ||
List< ManualResetEvent > | rgGradientReadyEvents, | ||
ManualResetEvent | evtAllCreated | ||
) |
The SolverInfo constructor.
rank0 | Specifies rank Solver that will run in the Worker. |
hSrcKernel | Specifies a handle to the kernel where the NCCL for this Solver was created (typically this is the kernel that also created the root Solver). |
hSrcNccl | Specifies the handle to the NCCL instance for this Solver (typically this is created on the kernel that also created the root Solver, and must be transferred to the kernel of the CudaDnn instance running in the Worker). |
nSolverRank | Specifies the rank of this Solver. |
nIterationOverride | Specifies the training iteration override to use. |
strCudaPath | Specifies the file path to the low-level CudaDnnDll.DLL file to use. Note, when null or emtpy, the path of the executing Assembly is used. |
rgGradientReadyEvents | Specifies the list of events used to coordinate with other Solvers. |
evtAllCreated | Specifies an event used to coordinate the creation of all participating Workers. |
Definition at line 545 of file Parallel.cs.
|
get |
Returns the event that is set after all Workers have been created.
Definition at line 624 of file Parallel.cs.
|
get |
Returns the file path to the low-level CudaDnnDll.DLL file to use. Note, when null or emtpy, the path of the executing Assembly is used.
Definition at line 568 of file Parallel.cs.
|
getset |
Returns the error (if any) that occured when running the solver thread.
Definition at line 640 of file Parallel.cs.
|
get |
Returns the event that is set when an error occurs.
Definition at line 649 of file Parallel.cs.
|
get |
Returns the event that is set after the gradients of the Solver in this Worker are ready.
Definition at line 632 of file Parallel.cs.
|
get |
Returns the event that is set after the Worker has completed initializing.
Definition at line 608 of file Parallel.cs.
|
get |
Returns the training iteration override to use.
Definition at line 576 of file Parallel.cs.
|
get |
Returns a handle to the kernel where the NCCL for this Solver was created (typically this is the kernel that also created the root Solver)
Definition at line 584 of file Parallel.cs.
|
get |
Returns the handle to the NCCL instance for this Solver (typically this is created on the kernel that also created the root Solver, and must be transferred to the kernel of the CudaDnn instance running in the Worker).
Definition at line 592 of file Parallel.cs.
|
get |
Returns rank Solver that will run in the Worker.
Definition at line 560 of file Parallel.cs.
|
get |
Returns the rank of this Solver.
Definition at line 600 of file Parallel.cs.
|
get |
Returns the event that is set after the Worker has started running.
Definition at line 616 of file Parallel.cs.